User:Io Katai/Khmer grammar

Khmer, like other Mon-Khmer languages, is an analytic language with Subject Verb Object (SVO) word order, in which syntax plays a more prominent role than morphology. The following article will attempt to provide a brief overview of both aspects of Khmer grammar.

Social Registers
Khmer employs a system of social registers (Reid, 1956) that designate a speakers's relative status within a conversation. Factors that influence this are age and social class (Royal, clerical, Laity). The effect of this register on lexical choice can be seen in selection of pronouns. Thus a young person speaking to an older person would use the polite 1psg ខ្ញុំ​​​ IPA /kʰɲɔm/ as opposed to អាញ IPA /æɲ/​, which would be used between peers. Register varies among the other persons and numbers as well and is modified by noun class (gender).

Affixation
There are contrasting theories on the number of infixes in Khmer and how they are formed at the morphophonological boundaries of syllables. These processes are fossilized from a productive derivational morphological system from Old and Middle Khmer (Jacob, 1963). Analysis shows that there were at least three infixation systems based on the morpheme /-m-/. According to Schiller (1994), infixes had three functions:

Nominalizing, Causative/Transitivizing, and Agentive.

Nominalizing infix, which used the nonsyllabic morpheme /-m-/ in Khmer as it is the most productive with over 220 derivatives (Jenner, 1969). The infix /-m-/ fits into the morphophonological environment of the verb that it nominalizes and the specific phonological conditioning of allomorphs in select Modern Khmer verbs to reveal patterns that apply to all verbs nominalized by this infix. Analysis of data will reveal the semantic relationship between the verb and its nominalized derivative as a pattern that will apply to all the pairs in this category. The data also show that the process is no longer productive and is a fossil from OKhm and MKhm.

Environments of the Infix The infix -m- is inserted at the first syllabic boundary of a polysyllabic verb. 1)	CV.CVC -> CV-m-.CVC

In the case that the verb is monosyllabic, the first consonant of the verb retains its inherent vowel, either High register /ɑː/ or Low register /ɔː/. High Register: 2)	CVC ->  C-ɑm.C-VC		ɗaə    	->	ɗɑm.naə   					Verb		Noun ‘walk’		‘travel’ ដើ		ដម្នើ			ដ = /ɗɑː/ Low Register: 3) 	CVC    C-ɔm.C-VC			kaət 	  	kɔm.naət Verb		Noun ‘be born’	‘origin’ គើត			គម្នើត		គ = /kɔː/

In examples 2 and 3 above the inherent vowel of each consonant becomes the nucleus of the first syllable to which the infix -m- is attached. As can be seen in example 2, ដ /ɗɑː/ has the inherent vowel /ɑ/ which is shortened when it becomes a closed syllable with the introduction of morph/phone [-m] as a coda. The same is evident for គ /kɔː/ in example 3. In the case of a consonantal cluster, the cluster is split by the morpheme /-m-/. The first consonant retains its inherent vowel, which is the nucleus of the first syllable and the second consonant becomes the onset of the second syllable. Consonantal Cluster 4)	C₁C₂VC   C₁V-m. C₂VC		criəŋ		cɑm.riəŋ										Verb		Noun						‘to sing’	‘song’						ច្រៀង		ចម្រៀង

In examples 1-4 an evident pattern emerges. The nonsyllabic infixational morpheme /-m-/, in every case, isolates the first consonant (C₁) of each word and becomes the coda of the first syllable of the word regardless of syllable structure. Stress is largely unaffected on the ultimate syllable of each word. The second consonant is displaced and becomes the onset of the following syllable. This pattern remains constant throughout all the examples in which the morpheme /-m-/ nominalizes Khmer verbs deriving nouns. 3.2 Phonological Conditioning  of Allomorphs In example 2) above, there is clearly some type of phonological conditioning occurring within the context of the utterance. It would seem that an -n- is inserted along with the known infix /-m-/.  This begs the question of whether or not this is a different affix altogether. Could it be /-mn-/ is another infix to consider? This is highly doubtful for three reasons.  First of all the function of the supposed second infix /-mn/ is exactly the same, nominalization is occurring and in the very same manner as /-m-/.  Second, the hypothetical infix /-mn-/ is split down the middle by a syllabic boundary /-m.n-/, which strongly indicates that this is not a single morpheme or a morph, but two separate units, an allomorph of the nonsyllabic infix /-m-/.  The third reason is that a phonological rule (see example 5) explains the insertion of /-n-/:		C		C		V		C			V 5)	Ø  [+nasal]   /    #	[+obstruent]	[+sonorant]	[+nasal]    $	_____[+sonorant] # [+coronal]	[+coronal]	[- front]	[+ labial]		[-back] [+voice]	[- high]				[-high] [-pulmonic]	[-long]					[-short] [+open] [+dipthong] 6)	Ø  /n/ 	/ ɗ ɑ m $ __ aə

Examples 5 and 6 show the context in which the phonologically conditioned allomorph /-m.n-/ is used. First, the utterance is monosyllabic and lacks a coda, leaving an open syllable. When the infix /-m-/ is introduced there is no consonant to use as the onset of the second syllable. Because the onset of the first syllable is [+coronal], the preceding consonant is a nasal bilabial and the last vowel is fronted, the natural phonological conditions are present to fill the gap with the +coronal nasal /n/. While sifting through the data for this project several instances occurred where there seemed to be another morpheme, but when these words were examined more closely it was found that the underlying form was still /-m-/. The following are some examples that were misleading at first glance. 7)	 tʰʊm 	 	tɔm.hʊm or tə̥m.hʊm Verb		Noun ‘grow’		‘size’	ធំ		ទំហំ

In example 7) above it would initially seem as though the aspiration /-h-/ was inserted due to some phonological process or that the underlying form is /m.h/. However, upon closer examination it is discovered that the aspirated onset of the first syllable /t/ has lost its aspiration.  This loss is reflected in the Khmer orthography the character ធ is pronounced [tʰɔ] while the character ទ is pronounced [tɔ].  The inherent aspiration /Cʰ/ on the initial consonant is treated by the infixation process like a consonantal cluster, such that the formula in example 4) can be applied to this infix. 7a)	C₁C₂VC 	 	C₁V-m. C₂VC	tʰʊm 		 	tɔm.hʊm

When the initial consonant is thought of as two phones rather than a single aspirated phone, the formula applies and the infixation process becomes typical. Though the process in example 7) that revealed its underlying form /-m-/ was somewhat easy to see, there were other examples that were slightly more difficult as was demonstrated in examples 2) and 5). In example, 8) below, the morph [-m-] seems to be missing altogether. This is another derivation that produces a phonologically conditioned allomorph. In this case, the allomorph is /-ŋ/.

8)	cʰkʌɦ 			cɑŋ.kʌɦ 	‘pick at’		‘chopsticks’	ឆ្កឹះ			ចង្កឹះ

It is likely that the original form for this was more like 8a) below:

8a)	cʰkʌɦ 			cɑm.kʌɦ

However, it is unnatural and awkward to pronounce [m.k] in example 8) above, the underlying form /-m-/ is assimilated by the velar stop [k]. Here the phonological formula is slightly easier to detect.  The environments for the phonological conditioning are very standard.  It is known that the underlying form is /-m-/, therefore:	/m/			/ŋ/			/k/ 8b)	[+nasal]		[+nasal]	/ ___ $	[+obstruent] [+bilabial]		[+velar]		[-continuant] [+velar] [-voice]

3.3 Semantic Relationship of Verbs and Derivations A survey of the lexical data in Jenner and Pou (1984) reveals an overwhelmingly obvious pattern in the semantic relationship between the verbs and their nominalized derivatives. Virtually all of the verbs that are nominalized with /-m-/ are actions that create something. Furthermore, the nominal lexemes derived from these verbs can be interpreted as the products of said actions. In review of the all the examples used in this paper this seems to be consistent as well: 2c) 	ɗaə   	 	ɗɑm.naə  	Verb		Noun ‘walk’		‘travel/journey’

In example 2) above, the verb is ‘walk’ and the noun is ‘journey.’ A journey is the natural and intuitive product of ‘walking.’  If one does enough walking he will have made a journey.

3c)	kaət 	 	kɔm.naət 	Verb		Noun ‘be born’	‘origin’

When a person is born somewhere this creates his/her origin. Once again this is a clear and obvious example. One more example should drive the point home: 4c)	criəŋ				cɑm.riəŋ							Verb				Noun ‘to sing’			‘song’

A song is obviously the product of singing. There are numerous examples; over 220 of (Jenner and Pou, 1984), but the point has been made almost to the point of redundancy. The semantic relationship is one of verb = action and the derived noun is what is produced from that action. 3.4 The Morpheme /-m-/ as a fossilized process Jenner and Pou (1981) Jacob (1963) all agree that there is currently no derivational morphology in Khmer, but that the systems in OKhm and MKhm were extremely rich and salient systems of affixation. According to current literature, Old Khmer is divided into Pre-Angkorian (611-801CE) and Angkorian (802-1431CE). Middle Khmer is considered to be from (1432-to Mid 18th Century). At some point during the 18th century, ostensibly there was a shift away from affixation and nominalized words were left frozen in the lexicon. Currently there are no sufficient theories to explain why this occurred. Suffice it to say that the process of infixation using any of the accepted morphemes cannot be applied to neologisms in Modern Khmer. Most neologisms in Modern Khmer are nouns. There are very few new verbs coined in Modern Khmer. For instance, a verb that may get borrowed into Khmer is ‘to drive.’  Instead of a new word for ‘driving’ the Khmer have adopted the use of the word for driving a horse cart ɓɒ. A verb in English such as ‘surf’ is not borrowed in as a verb but combined with an auxiliary verb to denote ‘to surf’: 9)	ciɦ rəlɔ:k	ride.wave	v. ‘to surf’

Without any new verbs, it is difficult to test if the /-m-/ morpheme can be inserted into a word to nominalize it. But even so, most old verbs cannot be nominalized using the infix. There is a finite number of verbs -223- that can accept the /-m-/ morpheme (Jenner and Pou, 1981). So, for the many reasons stated here, the process of infixation is no longer productive.

Pronouns

 * 1sg


 * 1pl

Attribution
In Khmer, the attribute is a unit used to modify the preceding head noun. The attribute may be either a noun, a verb, an indefinite or question word, a phrase, or an entire clause.


 * {| cellspacing="7" style="line-height: 1em;"


 * - style="font-weight: bold;"
 * pləv
 * dək
 * tumnɨɲ
 * - style="font-style: italic;"
 * road
 * transport
 * merchandise
 * - style="font-variant: small-caps;"
 * style="background: #bbbbff;" | head noun
 * colspan="2" style="background: #ffbbbb;" | attributive phrase
 * colspan="6" | 'road for transporting goods'
 * }
 * }

Additionally, the Khmer language uses attribution to indicate a possessive relationship between a pronoun and a noun.
 * {| cellspacing="7" style="line-height: 1em;"

However, quantifiers are used before the noun they modify.
 * - style="font-weight: bold;"
 * kaŋ
 * kɲom
 * - style="font-style: italic;"
 * car
 * style="font-variant: small-caps;" | 1.pronoun
 * - style="font-variant: small-caps;"
 * style="background: #bbbbff;" | head noun
 * style="background: #ffbbbb;" | attributive noun
 * colspan="6" | 'my car'
 * }
 * }
 * {| cellspacing="7" style="line-height: 1em;"


 * - style="font-weight: bold;"
 * sɑp
 * tŋay
 * - style="font-style: italic;"
 * every
 * day
 * - style="font-variant: small-caps;"
 * style="background: #ffbbbb;" | quantifier
 * style="background: #bbbbff;" | head noun
 * colspan="6" | 'every day'
 * }
 * }

Numeration
Numbers follow the word they modify, with an optional use of classifiers (function = predicate nouns).
 * {| cellspacing="7" style="line-height: 1em;"


 * - style="font-weight: bold;"
 * bɑɑŋ
 * pii
 * neak
 * - style="font-style: italic;"
 * older sibling
 * two
 * (person)
 * - style="font-variant: small-caps;"
 * style="background: #bbbbff;" | head noun
 * style="background: #ffbbbb;" | number
 * style="background: #ee8888;" | classifier
 * colspan="3" | 'two older siblings'
 * }
 * }

Reference pages

 * Inuit grammar
 * Vietnamese syntax
 * Category:Grammars of specific languages